NLP researcher at Ain Shams University: Arabic dialect speech benchmarks and multi-turn medical AI evaluation.
2
papers here
2
reviewed by Evolve
1
first-author
0
corresponding
—
works · OpenAlex
0
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-15) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.
Research profile
Output—
Not yet scored: the snapshot contains only a single-year work count (14 works recorded for 2026), with no multi-year publication series. This axis would measure cumulative output volume and its growth over the researcher's career.
Impact—
Not yet scored: citation coverage is insufficient — h-index and i10-index are recorded as 0, consistent with missing citation data rather than measured influence. This axis would measure citation impact and scholarly uptake of the work.
Focus0/100
Scores 0 because work is evenly split across two field clusters — 'speech processing / computational linguistics (NLP)' at 50% and 'natural language processing / medical AI evaluation' at 50% — and spreads across five topic clusters: Topic Modeling (10), NLP Techniques (4), Information Retrieval and Search Behavior (3), Multimodal Machine Learning Applications (3), and Machine Learning in Healthcare (3). This breadth across benchmark-building themes currently reads as low concentration.
Collaboration67/100
Scores 67 on a record of 28 distinct coauthors and a recurring circle of 12 frequent collaborators, including Nizar Habash, AbdelRahim Elmadany, Yannick Estève, Fethi Bougares and Mohamed Elhoseiny. The pattern is typical of shared-task and benchmark teams: the NADI 2026 overview paper is a middle-position contribution to a large multi-author consortium, while MedRoundsQA is a first-authored paper.
Activity79/100
Scores 79, driven by 14 works recorded in 2026 — the only year in the counts-by-year series — matching the recent-three-year total of 14. Output is therefore concentrated entirely in the current year, indicating a highly active recent phase whose longer-term trajectory cannot yet be assessed.
Reproducibility55/100
Scores 55 as the mean readiness across the 2 papers Evolve has reviewed, both rated medium. NADI 2026 is credited with honestly reported, realism-first evaluation across 8 Arabic speech subtasks, though TTS/SLT evidence is thin and rankings lack statistical tests; MedRoundsQA builds controlled multi-turn consultations from 1,387 board cases but depends on LLM judges.
Research summary
Youssef Mohamed is a researcher affiliated with Ain Shams University (Egypt) whose recorded work sits at the intersection of Arabic speech processing and the evaluation of natural language processing in medicine. The profile is name-resolved from OpenAlex bibliographic data; 'Youssef Mohamed' is a common name and no ORCID is linked, so identity confidence is moderate. Two Evolve-reviewed papers anchor the profile. As a middle-position co-author on 'NADI 2026: The Second Multidialectal Arabic Speech Processing Shared Task', he contributes to an 8-subtask benchmark built around realism-first evaluation of Arabic dialect identification, ASR, code-switching and text-to-speech. As first author of 'MedRoundsQA', he leads a persona- and difficulty-aware evaluation that converts 1,387 board-style cases into controlled multi-turn medical consultations, using counterfactual patient personas to isolate communication effects. The activity score of 79 reflects 14 works recorded in 2026 — the only year in the counts-by-year series — matching a recent-three-year total of 14. Collaboration scores 67, with 28 distinct coauthors and 12 recurring collaborators spanning Arabic NLP and speech groups. Topic clustering is led by Topic Modeling (10), followed by NLP techniques, information retrieval, multimodal applications and healthcare ML. Output and impact are not scored because citation coverage is insufficient; both reviewed papers carry medium reproducibility readiness (mean 55).
Collaboration pattern
With 28 distinct coauthors recorded, Youssef Mohamed works mainly in consortium-style benchmark teams. A recurring circle of 12 frequent collaborators spans Arabic NLP and speech groups — among them Nizar Habash, AbdelRahim Elmadany, Yannick Estève, Fethi Bougares, Salima Mdhaffar, Hamzah Luqman and Mohamed Elhoseiny — consistent with the multi-institution NADI shared-task effort, alongside a first-authored medical-AI evaluation line.
Research keywords
Highlights
NADI 2026: co-authoring the second multidialectal Arabic speech shared task
MedRoundsQA: first-authoring a persona- and difficulty-aware medical consultation benchmark
A benchmark-builder profile bridging Arabic speech and medical AI evaluation
The two reviewed 2026 papers split evenly between speech processing/computational linguistics and NLP/medical AI evaluation, with Topic Modeling the largest topic cluster (10) — a breadth that currently drives the focus score to 0.
Timeline
- 2026Co-authors (middle position) the NADI 2026 shared task overview covering 8 Arabic speech subtasks.
- 2026First-authors MedRoundsQA, converting 1,387 board cases into controlled multi-turn medical consultations.
- 2026OpenAlex records 14 works for 2026 — the only year in the counts-by-year series and equal to the recent-three-year total.
Papers in the Evolve catalog
Frequent co-authors
About this profile
This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.
- Bibliometric figures come from OpenAlex only; the profile is name-resolved from public bibliographic data with no ORCID linked, and 'Youssef Mohamed' is a common name, so some records may reflect namesakes.
- Output and impact are not scored: citation coverage is insufficient, and h-index and i10-index are recorded as 0, consistent with missing citation data rather than measured zero influence.
- The evidence base is two Evolve-reviewed 2026 arXiv preprints; reproducibility (mean 55) and field balance may shift as more work is reviewed.
- Counts-by-year data contains only 2026, so career-length productivity and trajectory claims cannot be made from this snapshot.